Download doyne lecture rhodopsin and autosomal dominant retinitis

DOYNE LECTURE
RHODOPSIN AND AUTOSOMAL DOMINANT RETINITIS
PIGMENTOSA
THADDEUS P. DRYJA
Boston, Massachusetts
lowe a debt of gratitude to the Master of the Oxford Con­
One approach to learning about the pathology and bio­
gress of Ophthalmology for the invitation to be here today.
logical mechanisms accountable for this disease is to
The Congress has been immensely exciting to me and has
provided me the opportunity to meet a number of distin­
guished British ophthalmologists whose work I have stud­
ied since specialising in ophthalmology 14 years ago. I
want to mention Alan Bird in particular, because his con­
tributions to our understanding of many hereditary retinal
diseases have been especially noteworthy. His expertise
encompasses most of the topic on which I will be speaking
today: retinitis pigmentosa. Here I would like to review
the approach my laboratory took to discover a gene
responsible for this devastating disease. This work has
held my attention for the last seven years.
Retinitis pigmentosa is the name given to a set of retinal
degenerations that have a number of clinical character­
istics in common (see Table I). Most cases, perhaps all, are
study the retina from the affected patients. Understandc
ably, most specimens donated for such research come
from deceased elderly individuals. The retinas of such
patients are typically severely degenerated and only an
occasional specimen will provide clues as to the patho­
genesis of the retinal degeneration.4-8 These reports, as
well as those describing retinas of the occasional younger
patient
demonstrated that families with X-linked retinitis pig­
mentosa can be further subdivided, since there are at least
two distinct loci on the X chromosome at which mutations
can cause the disease.1 It is likely that gene locus hetero­
geneity is also a feature of the autosomal dominant and
autosomal recessive forms of retinitis pigmentosa. In
some families with recessive disease, congenital or
acquired deafness can be a feature, in which case the diag­
nosis is more appropriately Usher's syndrome type I or
type II, respectively.2 There is recent evidence pointing to
a gene on chromsome 1 q as the cause of Usher's syndrome
type 11.3 The gene or genes responsible for Usher's syn­
drome type I, as well as the genes responsible for other
forms of recessive retinitis pigmentosa are somewhere
else in the human genome. In fact, there may be dozens of
genetic
loci
where
mutations
can
cause
retinitis
pigmentosa.
Correspondence to: Thaddeus P. Dryja, MD, Massachusetts Ey e and
Ear Infirmary, 243 Charles Street, Boston, MA 02114, USA.
Eye (1992) 6, 1-10
for research
post­
A few biochemical analyses have been performed on these
rare early cases, but thus far the data do not allow a
generally accepted consensus as to the mechanisms
accounting for the photoreceptor degeneration.
In view of the numerous patients with the disease
disease is inherited as an autosomal dominant trait in some
trait in still others. Furthermore, linkage studies have
eyes are donated
photoreceptors and/or the retinal pigment epithelial cells.
hereditary. The genetics of this disease is not simple. The
families, autosomal recessive in others, and as an X -linked
whose
mortem,9.10 reveal that the earliest affected cells are the
Table I.
I.
CLINICAL FEATURES OF RETINITIS PIGMENTOSA
A. Symptoms
1. Night blindness
2. Early loss of peripheral visual field
3. Late loss of central field as welJ
B. Signs
1. Pallid optic nerve head
2. Attenuated retinal vessels
3. Bone spicule pigmentary deposits in the periphery
4. Posterior subcapsular cataract
C. Electroretinographic abnormalities
1. Reduced amplitude of scotopic and photopic b-wave
2. Delay in time between flash of light and peak of b-wave
(delayed implicit time)
II. DISTRIBUTION OF CASES ACCORDING TO GENETIC
TYPE (based on ref. 12)
A.
B.
C.
D.
Autosomal dominant---43%
Autosomal recessive-20%
X-linked-8%
'Isolate' (i.e. only one affected family member, possibly
representing autosomal recessive disease, but could also be
new dominant or X chromosome mutation)-23%
E. Undetermined (i.e. adopted, uncertain family history, etc.)--6%
2
T. P. DRYJA
(50,000 to tOO,OOO in the United States alone)ll·12 as well
as its hereditary nature, it would be reasonable to try mol­
Examples of successes with the technique are the identi­
fication of genes causing chronic granulomatous disease,13
ecular genetics approaches to discover the relevant genes
Duchenne's muscular dystrophy,14 cystic fibrosis, 15-17, and
and consequently the biochemical defects. Genetic analy­
retinoblastoma. IS The amount of effort required can some­
sis of affected patients does not require retinal tissue, since
times be substantial: the approach still has not provided a
essentially the same genes are present in all somatic cells.
gene for Huntington's chorea despite over a decade of
Satisfyingly, the procurement of blood samples for the
work by many research groups, and despite the fact that
purification of leucoctye DNA is simple. The question
the chromosomal location of the responsible gene was dis­
becomes how to identify a gene that causes retinitis pig­
covered about eight years ago.19
ent in the human genome.
approach', has the advantage of being more straightfor­
mentosa among the 50,000 to 100,000 genes that are pres­
A molecular geneticist can take either of two major
routes to achieve this end
An alternative approach, called the 'candidate gene
ward but the disadvantage of being less assured of success.
(see Table II). The first
The investigator selects genes specifically expressed in the
approach, called the 'linkage' or the 'reverse genetics'
diseased tissue or that are known to code for proteins with
approach, depends upon finding a genetic marker in the
important functions in that tissue. Patients with the disease
human genome that is close to the gene of interest. For this
are then screened for mutations in each of those genes.
approach to be practical, it is necessary to have available
There are benefits and drawbacks of this conceptually
for study one or more large kindreds with scores of
simple approach. One advantage is that the method works
affected and unaffected members. After collecting blood
even if there is genetic heterogeneity, since it is necessary
samples and purifying DNA from each available family
that only some patients in the group under inspection have
member, the investigator analyses one chromosomal
a defect in the selected candidate gene. Another advantage
marker after another until he or she finds a marker whose
pertains to the fact that the candidate genes are selected
inheritance correlates with the inheritance of the disease
because of the known function of their protein products.
trait. Such a positive result indicates that the marker locus
Consequently, the discovery of a defect in such a gene
and the disease locus are in close proximity on the same
immediately provides insights into the pathophysiology of
chromosome. Since it is easy to ascertain the chromoso­
the disease. A disadvantage of this method is that it is
mal location of a marker, the scientist will soon deduce the
possible that no patients under study have disease due to
approximate location of the gene of interest. Fragments of
the candidate gene or genes that one chooses to examine.
DNA from that chromosomal region are cloned, and ulti­
This might be the situation if mutations of the tested can­
mately the investigator aims at finding a DNA sequence
didate gene are rare and no patients with them are included
that is expressed (i.e., is part of a gene) and is mutant in
in the laboratory's collection. Alternatively, a negative
affected individuals. The final task is to discover what the
result might only be due to imperfections in the techniques
identified gene does and why defects in it are pathogenic.
for finding the mutations, i.e. one might overlook the
The reverse genetics approach is generally expensive,
responsible defects. Finally, the reasoning that is used to
labour-intensive, and time-consuming. It has the advan­
select a candidate gene might be faulty; e.g. perhaps the
tage that it is almost sure to succeed given enough effort
gene is essential to life and that mutations are lethal.
and provided that a sufficiently large family is available
My laboratory has pursued the second approach in part
for study. It also will work if the same gene is known to
because I was fortunate to have the close collaboration of
cause disease in sets of small families under study.
Table II. Approaches to the identification of a disease gene without
prior knowledge of underlying biochemical defect
I.
LINKAGE
A. Collect leucocyte DNA from members of large families with
the disease.
B. Find a chromosomal marker that is co-inherited with the
disease. If such is found. then one knows that the disease gene
is 'close' to the marker locus.
C. Clone DNA fragments from the identified chromosomal
region.
D. Find sequences conserved during evolution, e.g., that are
similar in primates and rodents.
E. Determine whether the conserved sequences are expressed in
relevant tissues. If so, clone the corresponding mRNA
(cDNA) sequence.
F. Search for mutations in the identified transcriptional unit in
patients with the- disease.
II.
CANDIDATE GENE
A. Collect leucocyte DNA from unrelated patients with genetic
disease.
B. Col!ect cloned genes specific for diseased tissue.
C. Search for mutations in those genes in the patients.
Professor Eliot Berson, who has a large practice devoted
to the diagnosis and care of patients with retinitis pig­
mentosa. Since 1984 we have collected blood samples and
purified DNA from hundreds of patients with either ret­
initis pigmentosa or other forms of hereditary retinal
degeneration. Among the over 3,000 patients who have
volunteered for our research, we have concentrated our
efforts on a subset of 600 patients with retinitis pig­
mentosa who have been followed annually by Dr. Berson
for six years or more. These patients have been subdivided
according to the inheritance pattern of the disease: auto­
somal dominant, autosomal recessive, X-linked, 'isolate',
or undetermined'. The 'isolate' cases are those with only
one affected family member; most of these patients prob­
ably have an autosomal recessive form of disease, but
some could be X-linked and others could represent new
dominant
mutation'/,.
The
'un.determined'
categot'j
includes patients with uncertain family history (e.g.
orphans or patients who had been adopted). Blood
samples from the relatives of some selected patients have
3
DOYNE LECTURE: RHODOPSIN AND AUTOSOMAL DOMINANT RETINITIS PIGMENTOSA
also been obtained, although no families large enough for
invariably present in all affected members and no
pan-genomic linkage study were ascertained.
unaffected members of a particular family, one would
This large set of patients fulfills one of the requirements
strongly suspect that the gene had a mutation that was
for the candidate gene approach. The second requirement,
causing the disease. However, we never found statistically
of course, is the availability of candidate genes ready for
significant coinheritance of a candidate gene with retinitis
analysis. Fortunately, this was no major hurdle. During the
pigmentosa in the few pedigrees that were analysed.
1980s genes for a number of important photoreceptor pro­
The second analysis we performed with RFLPs was
teins had been cloned, including rhodopsin,20 interphoto­
based on the concept of linkage disequilibrium. Alleles of
receptor
RFLPs are typically found to be distributed among indi­
retinoid
binding
protein,21,22
cellular
retinaldehyde binding protein,z3 48K protein (also called
viduals according to the Hardy-Weinberg equilibrium (see
arrestin or S-antigen),24 the alpha subunits of rod and cone
Table III). Departures from Hardy-Weinberg equilibrium
transducin,25 the gamma subunit of cGMP-phosphodieste­
can indicate a bias in the selection of individuals in the set
rase,26,27 etc. Most of these genes were known to be
under study. One explanation for such a bias is that a pro­
expressed only in retina, and their protein products were
portion of the individuals in the set descend from a com­
receptors. The only issue in my mind was how often (not
overrepresentation of an RFLP allele that was carried by
whether) these genes were mutant in patients with heredi­
that ancestor. If such a result were found among a large set
tary photoreceptor dysfunction.
of supposedly unrelated patients with, say, autosomal
considered to be essential to the functioning of photo­
Since those patients with a defect in one of these can­
mon
ancestor,
in
which
case
there
would
be
an
recessive retinitis pigmentosa, one would have suggestive
didate genes might represent only a small minority of the
evidence that the overrepresented allele carries a mutation
cases with a given type of retinitis pigmentosa (or even
that was carried by this presumed distant ancestor. At that
some other hereditary retinal disease), the key was to
point one could focus time-consuming techniques of
devise methods that could distinguish those individuals
obtaining the DNA sequence of the gene to those individ­
among the hundreds of patients who were available for
uals with the overrepresented allele. With this reasoning in
study. When the actual searching for mutations in these
mind, we used cloned candidate genes with known RFLPs
genes was initiated around 1987, the only method avail­
to look for departures from Hardy-Weinberg equilibrium
able for quickly screening for mutations was the Southern
among our sets of patients with various forms of retinitis
blot technique. This method is excellent for detecting
pigmentosa; we found none.
deletions, insertions, and other gene rearrangements, but it
has the drawback that it misses most point mutations.
Over the next few years, members of my laboratory
In spite of these persistently negative results, we were
confident that the approach was a sound one and we per­
severed with it. A milestone in this work occurred in 1989,
used the Southern blot technique to search for mutations in
but not in our laboratory. Dr. Peter Humphries in Dublin,
some candidate genes in our' core' set of 600 patients with
Ireland, had been using the linkage approach in his studies
retinitis pigmentosa. Despite years of work, we found no
of a large Irish pedigree with autosomal dominant retinitis
mutations with this relatively insensitive technique in the
pigmentosa. This was the first approach discussed above;
genes coding for rhodopsin, interphotoreceptor retinoid
the one we had not taken. Dr. Humphries announced in
binding protein,28 cellular retinaldehyde binding protein,29
August that he had discovered a marker that was closely
48K protein,z8 the alpha subunits of the rod and cone28
Table III.
transducins, or the gamma subunit of phosphodiesterase.30
Realising that we could be missing point mutations, which
were at that time detectable only after a tedious, time­
consuming process, we simultaneously took advantage of
quicker, indirect methods that could possibly suggest the
RFLP alleles
at the
hypothetical
test locus
existence of mutations in our candidate genes. These
indirect methods utilise RFLPs, which are naturally
occurring variations in the DNA sequence of genes. We
located RFLPs in the genes coding for cellular retinalde­
hyde binding protein,29 48K protein,28 the alpha subunit of
cone transducin,28 and the gamma subunit of phosphodies­
terase,30 among others. RFLPs in some candidate genes,
such as the gene for interphotoreceptor retinoid binding
protein, had been discovered by other groupS.31,32 These
RFLPs were used two ways in our studies.
First, although RFLPs in themselves do not ordinarily
confer any particular phenotype, they allow one to trace
the inheritance of candidate genes through a family to see
if any are coinherited along with retinitis pigmentosa. If a
specific copy of a gene, identified by its RFLP alleles, was
Hypothetical example of linkage disequilibrium
1,1
1,2
2,2
Totals
Number of
control
subjects with
given RFLP
alleles
among
group of 100
25
50
25
100
100 patients with recessive retinitis
pigmentosa, 20 of whom descend from
a shared ancestor with the '1' allele at
the test locus
80 not from
ancestor
20 from
ancestor
20
40
20
20
0
0
Sum
40
40
20
100
According to Hardy-Weinberg equilibrium if two codominant alleles
are in a 50:50 proportion in a population, then the genotypes among a
group of 100 individuals from that population should approximate the
numbers given above. In contrast, the 100 patients with autosomal
recessive retinitis pigmentosa do not have this distribution because 20 of
them descend from a single ancestor who had the '1' allele at the test
locus and who had a mutation causing the disease in that allele. These 20
patients are all homozygous for the mutation causing retinitis pig­
mentosa and consequently are '1,1' homozygotes for the RFLP. A statis­
tical test such as Chi-square will provide the likelihood that the
differences in the two groups are statistically significant.
4
T. P. DRYJA
linked to the disease gene in this family. Since the marker
No unaffected individual has been found to carry any of
was derived from the long arm of chromosome 3, the
these mutations. Combining all of this data, it appears that
disease gene must be on that same chromosome arm.
mutations in the rhodopsin gene are the cause of about
When we heard this news a specific candidate gene came
25-30% of cases of dominant retinitis pigmentosa. The
to mind because it was known to be on the same chro­
remaining cases are due to defects at other loci, and the
mosome arm: the rhodopsin gene.33,34 Until that time, we
search for those loci is understandably under active
had done little work with the rhodopsin gene because no
investigation.
informative RFLPs were known to be at the locus. We had
Faced with a sudden abundance of new data about a
done some Southern blot studies to look for gene deletions
disease, one strives to make sense of it and to organise it in
or rearrangements and had found none among over 100
a rational manner. There are a variety of approaches one
patients whom we analysed (unpublished results). After
can take to analyse this data. Now I will consider what this
learning about Humphries' results, however, we suspen­
set of mutations tells us about the mutability of the rho­
ded work on other candidate genes and devoted the major
dopsin gene. I will speculate on the possible pathogenic
portion of the laboratory 's resources to searching for point
properties of the mutant rhodopsin molecules that are
mutations in the rhodopsin gene in our patients with auto­
encoded by these abnormal alleles. Finally, I will review
somal dominant retinitis pigmentosa.
some of the clinical characteristics of the patients who
Since we already knew that deletions of the gene were
carry these mutations.
not present in our patients, we decided to use advanced
Mutations are of fundamental importance to genetics,
methods based on the technique called the polymerase
and the subject would be quite boring without them.
chain reaction (PCR) for rapidly sequencing DNA from a
Hence, geneticists have developed categorisation schemes
specified gene. We had previously gained some experi­
for them. Mutations can affect a single base pair (point
ence with this technique from our efforts at finding point
mutations), or many base pairs. They can be deletions
mutations in the retinoblastoma gene in patients with ret­
ranging in size between one base pair and millions of base
inoblastoma.35,36 The application of the method to the rho­
pairs. They can be insertions of DNA from another locus,
dopsin gene was facilitated by the fact that Dr. Jeremy
insertions or duplications of DNA from the same locus, or
Nathans had already determined the gene's complete
inversions, translocation, etc. The mutations so far found
DNA sequence?O A map of the gene, based on Nathans'
in the DNA sequence of the rhodopsin gene are almost all
results, is shown in Figure 1. We developed a protocol for
of one category: point mutations. As shown in Table IV,
directly sequencing the coding sequence from the rho­
they are more frequently transitions (substitution of a
dopsin gene and selected 20 unrelated patients with auto­
purine base for another purine, or a pyrimidine base for
somal dominant retinitis pigmentosa for this intensive
another pyrimidine) rather than transversions (substitution
analysis. Within six weeks of learning of Humphries' find­
of a purine for a py rimidine or vice versa. Among the tran­
igns, a research assistant in my laboratory, Terri McGee,
sitions, the changes C-to-T and G-to-A (which are really
had discovered the same point mutation in five of those
the same mutation, the difference depending on whether
patients.37 This mutation changed a cytosine to an adenine
one reads the sense or antisense strand of DNA) are the
'C-to-A transversion') in codon 23. This codon
most frequent. Is this preference for these two related tran­
normally specifies proline as the 23rd amino acid in the
sitions (among the twelve possible single base substitu­
sequence of human rhodopsin; with the C-to-A mutation
tions) specific to the rhodopsin gene, possibly telling us
the codon would instead specify histidine.
something about retinitis pigmentosa or the mutability of
(a
We sought additional evidence that this alteration in the
this gene? Probably not, since this bias for transitions, and
DNA sequence was a cause of retinitis pigmentosa.
especially the C-to-T and G-to-A transitions, is a feature
Further testing revealed that about 10% of our patients
of the point mutations found at many loci. It may relate to
with autosomal dominant retinitis pigmentosa carried this
the as yet poorly defined mechanisms responsible for
mutation; none of over 100 unrelated, unaffected individ­
germline mutations in humans.
uals did. In a few families, we were able to trace the inher­
Another interesting feature of these mutations is their
itance of this mutation through three generations; it
rarity. Most of them are found in only one family, indi­
perfectly correlated with the inheritance of retinitis pig­
cating that many of them might have arisen in a single
mentosa. Finally, it was likely that the proline at position
ancestor of each family. The first mutation we discovered,
23 was important to the structure or function of rhodopsin
the Pr023His mutation, is an extreme example of this. The
since that amino acid has not changed during evolution
17 'unrelated' families that we have described with this
among the vertebrate rod and cone opsins.38-4o
mutation all carry the same rare marker at a microsatellite
This discovery of a cause for retinitis pigmentosa
polymorphism within the first intron of the rhodopsin
prompted our group and other groups to search for other
gene,41 in addition to the C-to-A transversion in codon
mutations in the rhodopsin gene. In all, over 30 distinct
23.42 It is more likely that the pro23His mutation arose
mutations have been discovered among patients with
only once on a copy of the rhodopsin gene with this
dominant retinities pigmentosa in North America, Europe,
uncommon microsatellite allele than many times but
and Japan (see Table IV ). In every family so far studied,
always on a rhodopsin allele that happened to have this
the mutation invariably was coinherited with the disease.
rare microsatellite sequence. In support of the idea that
DOYNE LECTURE: RHODOPSIN AND AUTOSOMAL DOMINANT RETINITIS PIGMENTOSA
5
MAP OF THE HUMAN RHODOPSIN GENE WITH
LOCATIONS OF DNA POLYMORPH ISMS AND SOME MUTATIONS
Pr0347Leu
Pr0347Ser
lhr58Arg
Pro23His
)(
,
Rsal
RFLP
(CA)n
microsatellite repeat
I
I
I
I
I
I
o
I
-mutations
I
I
I
I
I
I
I
I
I
I
I
I
1,000
I
I
I
I
I
I
I
2,000
I
I
I
I
I
I
I
3,000
I
I
I
I
I
I
I
I
I
I
4,000
C-to-A
-polymorphisms
rare variant
I
I
I
I
I
I
I
I
I
5,000
I
I
I
I
I
I
I
I
I
I
6,000
I
I
I
I
I
I
I
I
I
7,000
Scale measuredin base pairs ofDNA
W·l l l-I·II
= untrans/ated region
o coding region
� numberofexon
=
=intron
Fig. I. Map of the human rhodopsin gene. To the left is the5' end ofthe gene; to the right is the3' end. The orientation of this gene with
regard to the centromereand telomere of chromosome3 is not yetknown. The map shows the positions of polymorphisms in the gene
thatare found in humans. Also shown are the approximate positions of some of the early mutations found among patients with dominant
retinitis pigmentosa.
these 'unrelated' familes share a common ancestor is the
fact that the Pro23His mutation has never been found in
Most of the mutations are missense mutations, i.e. they
would be expected to cause a substitution of one amino
Europe or Asia. Coupling that information with the ances­
acid for another in rhodopsin. A few of them are deletions
try from pre-revolutionary settlers to North America that
or point mutations that would result in the removal of one
some of these 17 families claim, it becomes clear that the
or a few amino acids in the protein. None of the mutations
founder of this mutuation might have been an early colo­
nist, possibly from Great Britain.
appear to be null mutations, i.e. mutations that would
result in little or no protein product. In view of this, it
A few of the mutations, however, definitely occurred
appears that the disease associated with these mutations is
more than once in human history. A C-to-T transition in
due to the production of a mutant form of rhodopsin that is
codon 347, changing the codon from specifying proline to
somehow toxic to photoreceptors.
specifying leucine (Pr0347Leu), is an example of this. We
Consequently, a tabulation of which regions of the pro­
have found the Pro347Leu mutation in eight unrelated
tein are affected by these amino acid substitutions might
families.42 Analysis of the microsatellite repeat polymor­
reveal insights as to the nature of this supposed toxicity.
phism in intron 1 indicates that there are at least two separ­
Figure 2 shows a schematic representation of the rho­
ate founders of this mutation among seven of these
dopsin molecule. According to current models, this pro­
families. Furthermore, the eighth family presented us with
tein traverses the disc membrane seven times. The amino
the only instance we could find of a new germline muta­
terminal end is in the intradiscal space, and the carboxy
tion in the rhodopsin gene. In this family the Pro347Leu
terminus is in the cytoplasm. The circles indicate amino
mutation was present in a patient and her child but not in
acids affected by the mutations seen in patients with auto­
her parents. This mutation has been found also in Great
somal dominant retinitis pigmentosa. Note that they could
Britain43 and in Japan,44 presumably having arisen in yet
other founders. The explanation for this relative 'hotspot'
for mutations in the rhodopsin gene might be that codon
347 is unusually susceptible to the obscure mechanisms
responsible for C-to-T transitions in the human germline.
be divided into three groups according to the location of
the affected amino acids in rhodopsin (see Table IV). In
the first group are mutations that affect amino acids in the
intradiscal space. Many of these affecting amino acids
near the cysteines involved in a disulfide bond connecting
6
T. P. DRYJA
Table IV. Mutations found ill the: rhodopsin gene in patients with
autosomal dominant retinitis pigmentosa
No.
Mutation
References
Transition/
transversion
the fact that rod photoreceptors normally synthesise a
large amount of rhodopsin. New molecules of rhodopsin
are
synthesised
daily
by
rods. Rhodopsin
actually
accounts for approximately 80% of all the protein in the
rod outer segment,46 and approximately 10% of the outer
Group I-mutations affecting amino acids in the intradiscal space
50--53,57
1. Thr17Met
transition
transversion
37,42,50--54,60,62,63
2. Pro23His
transition
51
3. Pro23Leu
4. Gly106Trp
52,53
transversion
5. CysllOTyr
[author's laboratory, unpublished transition
transition
6. Tyr178Cys
52,64
7. Glul81Lys
51
transition
8. Gly182Ser
50
transition
9. Ser186Pro
51
transition
10. Gly188Arg
51
transition
11. Asp190Asn
51,65
transition
12. Asp190Gly
51-53
transition
segment is renewed each day.47 Now recall that ' old' rho­
Group II-mutations affecting amino acids in transmembrane
domains
13. Phe45Leu
52,53
transition
14. Gly51Arg
[author's laboratory, unpublished] transversion
15. Gly51 Val
51
transition
16. Thr58Arg
42,50--53,57,58,61
transversion
52,53
17. Val87Asp
transversion
18. Gly89Asp
transition
51-53
19. Leul25Arg
51
transversion
20. Arg135Leu
2 transversions
52,53,57
21. Arg135Trp
52,53,57
transition
51
22. Cys167Arg
transition
23. Pro171Leu
51
transition
24. His211Pro
65
transversion
25. Ile255Dei
neither
43,66
26. Pro267Leu
transition
50
27. Lys296Glu
65
transition
transportable to the outer segment disc membrane, but
Group III-mutations affecting amino acids in the cytoplasm
28. Del68-71
65
neither
29. Gln344Ter
52,53,57
transition
30. Va1345Met
51,59
transition
31. Pro347Arg
67
transversion
32. pro347Leu
42-44,51-53, 56
transition
42,51
33. Pro347Ser
transition
the first and second intradiscal loops. In the second group
are mutations affecting amino acids in the transmembrane
regions. Many of these mutations replace a hydrophobic
amino acid with a charged one. The third group has the
few mutations that affect amino acids in the cytoplasmic
regions of the protein. Most of these affect the last few
residues at the cytoplasmic end of the molecule.
Most of the mutations in the first two groups probably
destroy the normal three dimensional conformation of
rhodopsin. This conjecture relies on evidence that the
intradiscal domains of rhodopsin are important in main­
taining the shape of rhodopsin.45 It is consistent with the
notion that adding charged amino acids to transmembrane
domains probably destabilises those domains. Further­
more, many of the mutations in these groups involve pro­
line residues, an amino acid that is important in protein
folding. The cluster of mutations affecting amino acids
near the disulfide bond connecting the first and second
intradiscal loops also conforms with this idea, since this
disulfide bond is also thought to be essential for a func­
tional conformation of rhodopsin.45
dopsin is not catabolised intracellularly by rods. Instead, it
is ingested daily be the neighbouring retinal pigment epi­
thelial cells as they consume the tips of the rod outer seg­
ments. The normal situation is therefore that rods
manufacture an abundance of a particular type of protein
that they are not required to recycle. Envision what would
occur if the rods could not utilise the retinal pigment epi­
thelial 'disposal site' for this protein. Mutant rhodopsin
molecules with improper conformation might not be
instead might accumulate in the inner segments or other
regions of the cell. In the framework of the model I pro­
pose, rods have no catabolic pathway to deal adequately
with this load of mutant rhodopsin molecules. The pre­
sumed build up of rhodopsin molecules in the rods is what
may lead to their demise.
Further support for this model comes from two sets of
experiments done by other groups. The first deals with the
glycosylation of rhodopsin. Carboyhydrate moieties are
normally covalently bound to two asparagine residues
near the amino terminus of the protein. This glycosylation
is probably important to the normal transport of rhodopsin
to the outer segment discs, because when rods are exposed
to tunicamycin, an inhibitor of glycosylation, rhodopsin
accumulates in the inner segment.48,49 One of the muta­
tions found in dominant retinitis pigmentosa alters a threo­
nine (at position 17) located two residues from one of the
normally glycosylated asparagines (at position IS-see
Figure 11).50-52 The mutation is referred to as ThrI7Met.
One requirement for glycosylation of an asparagine is that
the amino acid two residues away be a threonine or a
serine. Since the Thr17Met converts the necessary threo­
nine at this position to a methionine, glycosylation of
asparagine-IS would be excluded. The results from the
experiments with tunicamycin suggest that this mutant
rhodopsin with defective glycosylation would accumulate
in the inner segment.
Other data supporting this theory comes from the work
of Nathans' group at Johns Hopkins.53 Wild-type and
mutant forms of rhodopsin were expressed in vitro using
COS cells. When wild-type rhodopsin is expressed, it is
detectable on the surface of these cells, consistent with its
expected affinity for cell membranes. However, most
mutant forms of rhodopsin found in patients with domi­
nant retinitis pigmentosa remain in the cytoplasm.
Again, the mutations in groups one and two (see Table
IV) generally conform with the theory that the mutant rho­
dopsin might be toxic to photoreceptors because they may
amass excessively in the cytoplasm. This explanation,
Evidently, rhodopsin molecules with improper confor­
although appealing, has a few weaknesses. First, it does
mation are toxic to photoreceptors; what could be the
not appear to explain the retinal degeneration associated
reason? The explanation most appealing to me relates to
with the mutant rhodopsins in group three, especially
DOYNE LECTURE: RHODOPSIN AND AUTOSOMAL DOMINANT RETINITIS PIGMENTOSA
7
AMINO ACIDS AFFECTED BY MUTATIONS
IN AUTOSOMAL DOMINANT RETINITIS PIGMENTOSA
��:
G
(f®A@3)S T ET K
S VT ASA
R L LP
Q Q QE
F
K
M NF RFGEAA
SATT Q
S
HKK
� Q
.r
p
A
K
E
Q
E
KI G
T
N
K
K
C
TV
v
N
H
cC'
PLN Vvvy
T
R
V
F
T
V
A
M
M
M
fi"\
Y
I
E<B:
L
I
I
Yv
Iy
IT'L LN L A I L G MV G QCy �VM I V
F
L
V
A V V L F AT F F I V A
L
P N
Outer
\.!..IN
. VW
I
lY
F
OA L�WS
P I@F
I
M A I
F
M LC I � AA
A
.oM
Segment
LI V �@L E1 G A L:© I P T VW
- A
v® �
G
A yQ32F
FFPA
LM® G
F
A
A
P
L
T
L
T
Membrane
S
A
P®
F
Y AA TSyL FF G
L
M vY FAVy TIF M
A
G
E
I
w
ML
T S
F I
P I_
- -'
L.:.;:::S
FV
FS
L NL
Q
©. I§'
�I SE
W
y GH G'
Q G SN
N
F
1
P T \c�cQ �
N
© L
<,9F'
A
vF@
v
I@
E
T
KP
Y
Y
L
Y Q P YEF@S R v vGG)AN-glycos.
Intradiscal space
M GT EGPNFYV PFS
N
glycos.
Cytoplasm
\!j
J
J
�
'--_----'
L..--�
�
Fig. 2. Schematic model of rhodopsin. The protein is composed of 348 amino acids in a linear array, shown using the standard single
letter code. The string of amino acids traverses the outer segment disc membrane seven times. The amino terminus of the protein is in
the intradiscal space; the carboxy terminus is in the cytoplasm. Numbering of amino acids begins at the methionine residue at the amino
terminus ('M' in the lower left hand corner of the figure.) The two glycosylation sites near the amino terminus are indicated with the
abbreviation 'glycos', and are at amino acids 2 and 15. The lysine (K) residue at position 296 to which 11-cis-retinal is covalently
bound is approximately in the middle of the seventh transmembrane domain (farthest one to the right). Circles indicate the amino acids
affected by mutations in patients with autosomal dominant retinitis pigmentosa.
those mutations affecting the carboxy end of the molecule.
symptom of retinitis pigmentosa, and by the observation
Such mutant rhodopsins correctly assemble in the cell
that electroretinograms (ERGs) show a greater reduction
membrane when they are expressed in COS cells.53 Per­
in the rod response compared with the cone response to
haps there is a signal sequence at this end of the molecule
flashes of light in early cases. In addition, the ERGs show
that is important for the intracellular transport of rho­
a pathological delay between the stimulating flash of light
dopsin, an idea put forward by Paul Hargrave. When a
and the rod response.37.42.54 One puzzling feature of rho­
mutation affects the signal sequence, rhodopsin might
dopsin-related dominant retinitis pigmentosa is that cones
accumulate in the cell body in a fashion similar to what I
also degenerate as the disease progresses. Why should a
propose for the other mutant rhodopsins. Another possible
defect in a rod-specific protein ultimately induce degener­
explanation is that the entire theory I have put forward
ation of cones as well? Perhaps the answer is a con­
here is mistaken; photoreceptor degeneration might be a
sequence of the small proportion of cones relative to rods
(5 million cones vs. 90 million rods).55
consequence of some other absent, vital property or newly
in the human retina
acquired, toxic property of the mutant rhodopsins.
The surrounding large-scale destruction of rods might
Do the ophthalmological findings of the patients who
produce an environment too hostile for the scattered
carry these mutations help in understanding these forms of
cones. However, the rod-specific nature of rhodopsin may
retinitis pigmentosa? Professor Eliot Berson, my close
provide a basis for future therapeutic approaches to this
collaborator in this work, has meticulously examined the
disease. If only one could devise a way to preserve cones,
patients in whom we have found mutations. In view of the
fact that rhodopsin is synthesised in the rods and not
cones, the retinal degeneration in young cases more
especially the macular cones, vision would be maintained.
Another question is whether the severity of the retinal
degeneration is a function of the specific mutation in the
severely affects rod rather than cone function.37.42.54 This is
rhodopsin gene a patient carries. It turns out that there is a
evident by the fact that nyctalopia is a frequent early
considerable amount of variation in the severity of retinitis
8
T. P. DRYJA
pigmentosa even among related patients with the same
Doyne Memorial Lecture. I trust that he will enjoy the
mutation. Despite this variation, the knowledge of which
hospitality and fellowship you have so kindly bestowed on
rhodopsin mutation a patient carries can have predictive
me.
value. For example, patients with the Pro23His mutation
generally
have
a
slower
course
than patients
with
Pr0347Leu mutation, with both a greater ERG signal and a
greater amount of remaining visual field at a given age.56
Most patients with Pro23His are expected to retain some
useful vision well into the seventh decade of life, whereas
patients with Pr0347Leu would be expected to be blind
many years earlier, on average. Too few patients have been
examined with some of the other mutations to make statis­
tically significant correlations regarding the clinical
course of retinal degeneration. Nevertheless, the varia­
tions in severity among the patients with different muta­
tions of the rhodopsin gene, reported by our group as well
as others,54,57-62 makes it probable that each rhodopsin
mutation will some day indicate to the ophthalmologist a
particular clinical course and a forecast of the age at which
a patient is most likely to lose all useful vision.
It is important to emphasise that not all patients with
autosomal dominant retinitis pigmentosa carry a mutation
in the rhodopsin gene. As mentioned earlier, only about 25
to 30% of cases are due to defects in this gene. The remain­
ing autosomal dominant cases, not to mention cases with
autosomal recessive and X-linked disease, develop retinal
degeneration due to mutations in other genes. Over the
next few years, I expect that many of these other genes will
be identified. At that time we should have a clearer picture
of the range of genetic defects that cause this disease. Any
properties that these genes or their protein products share
might be clues to understanding the mechanisms for her­
editary retinal degeneration. Hopefully, this knowledge
will help in finding a therapy that can slow or stop the pro­
gressive loss of vision characteristic of all forms of ret­
initis pigmentosa.
In this lecture I have recounted in a historical fashion
the approach that lead to the discovery that defects in the
rhodopsin gene cause some forms of dominant retinitis
pigmentosa. Molecular genetics techniques are extremely
powerful and still improving. They are becoming easier
and cheaper to perform and more widespread in their
application. We should expect advances in our under­
standing of many hereditary eye diseases during our life­
times. I especially await the identification of the gene
causing the hereditary retinal disease that bears Professor
Doyne's name, i.e. Doyne's honeycomb choroiditis. As an
autosomal dominant condition, it should be straightfor­
ward for an interested ophthalmologist to collect blood
samples from one or more large families with the disease
and use either the linkage approach or the candidate gene
approach to identify the responsible locus. Such a study
might provide a fundamental insight into age-related mac­
ular degeneration, a common disease of the elderly for
which we know too little about the pathogenesis. I predict
that in the next 20 years the 'Doyne's gene' will be iso­
lated. The responsible investigator will no doubt be
honoured by an invitation to this Congress to present the
REFERENCES
1. Chen JD, Halliday F, Keith G, Sheffield L, Dickinson P,
Gray R, Constable I, Denton M: Linkage heterogeneity
between X-linked retinitis pigmentosa and a map of 10
RFLP loci. AmJ Hum Genet 1989,45: 401-11.
2. Kimberling Wl,Moller CG,Davenport SLH,Lund G,Gris­
som TJ,Pri1uck I,White V,Weston MD,Biscone-Halterman
K, Brookhouser PE: Usher syndrome: clinical findings and
gene localisation studies. Laryngoscope 1988,99: 66-72.
3. Kimberling Wl, Weston MD, Moller C, Davenport SL,
Shugart Y Y, Pri1uck lA, Martini A, Milani M, Smith Rl:
Localisation of Usher syndrome type II to chromosome 1q.
Genomics 1990,7: 245-9.
4. Kolb,l and Gouras P: Electron microscopic observations of
human retinitis pigmentosa, dominantly inherited. Invest
Ophthalmol Vis Sci 1974,13: 487-98.
5. Verhoeff FH: Microscopic observations in a case of retinitis
pigmentosa. Arch Ophthalmol 1931,5: 392-407.
6. Friedenwald 1: Discussion of Verhoeff's observations of
pathology of retinitis pigmentosa. Arch Ophthalmol 1930,
4: 767-70.
7. Cogan DG: Pathology [of retinitis pigmentosa]. Trans Am
Acad Ophthalmol Otolaryngol 1950, 54: 629-61.
8. Szamier RB and Berson EL: Histopathologic study of an
unusual form of retinitis pigmentosa. Invest Ophthal Vis Sci
1982,22: 559-70.
9. Flannery lG, Farber DB, Bird AC, Bok D: Degenerative
changes in a retina affected with autosomal dominant ret­
initis pigmentosa. Invest Ophthalmol Vis Sci 1989, 30:
191-211.
10. Szamier RB,Berson EL,Klein R,Meyers S: Sex-linked ret­
initis pigmentosa: ultrastructure of photoreceptors and pig­
ment epithelium. Invest Opthalmol Vis Sci 1979, 18:
145-60.
11. Boughman lA,Conneally PM,Nance W E: Population stud­
ies of retinitis pigmentosa. Am J Hum Genet 1980, 32:
223-35.
12. Bunker CH,Berson EL,Bromley WC,Hayes RP,Roderick
T H: Prevalence of retinitis pigmentosa in Maine. AmJ Oph­
thalmol 1984, 97: 357-65.
13. Royer-Pokora B, Kunkel 1M, Monaco AP, Goff SC, New­
burger PE, Baehner RL, Cole FS, Curnutte JT, Orkin SH:
Cloning the gene for an inherited human disorder-chronic
granulomatous disease-on the basis of its chromosomal
location. Nature 1986,322: 32-8.
14. Monaco AP, Neve RL, Colletti-Feener C, Bertelson Cl,
Kurnit DM, Kunkel 1M: Isolation of candidate cDNAs for
portions of the Duchenne muscular dystrophy gene. Nature
1986,323: 646-50.
15. Rommens 1M, Iannuzzi MC, Kerem B, Drumm ML, Mel­
mer G,Dean M,Rozmahel R,Cole lL,Kennedy D, Hidaka
N,Zsiga M,Buchwald M,Riordan lR,Tsui LC,Collins FS:
Identification of the cystic fibrosis gene: chromosome walk­
ing and jumping. Science 1989,245: 1059-65.
16. Riordan lR,Rommens 1M,Kerem B,Alon N,Rozmahel R,
Grzelczak Z, Zielenski 1, Lok S, Plavsic N, Chou lL,
Drumm ML, Iannuzzi MC, Collins FS, Tsui LC: Identi­
fication of the cystic fibrosis gene; cloning and charac­
terisation of complementary DNA. Science 1989, 245:
1066-73.
17. Kerem B,Rommens 1M,Buchanan lA,Markiewicz D,Cox
TK,Chakravarti A,Buchwald M,Tsui LC: Identification of
the cystic fibrosis gene: genetic analysis. Science1989,245:
1073-80.
18. Friend SH, Bernards R, Rogelj S, Weinberg RA, Rapaport
DOYNE LECTURE: RHODOPSIN AND AUTOSOMAL DOMINANT RETINITIS PIGMENTOSA
JM, Albert DM, Dryja TP: A human DNA segment with
properties of the gene that predisposes to retinoblastoma and
osteosarcoma. Nature1986,323: 643-6.
19. Gusella JF, Wexler NS, Conneally PM, Naylor SL, Ander­
son MA, Tanzi RE, Watkins PC, Ottina K, Wallace MR,
Sakaguchi AY, et al: A polymorphic DNA marker gen­
etically linked to Huntingdon 's disease. Nature 1983,306:
234--8.
20. Nathans J and Hogness DS: Isolation and nucleotide
sequence of the gene encoding human rhodopsin. ProcNatl
Acad Sci USA1984,81: 4851-5.
21. Si JSS, Borst DE, Redmond T M, Nickerson JM: Cloning of
cDNAs encoding human interphotoreceptor retinoid-bind­
ing protein (IRBP) and comparison with bovine IRBP
sequences. Gene1989,80: 99-108.
22. Fong SL, Fong WB, Morris TA, Kedzie KM, Bridges CDB:
Characterisation and comparative structural features of the
gene for human interstitial retinol-binding protein. J Bioi
Chem 1990,265: 3648-53.
23. Crabb JW, Goldflam S, Harris SE, Saari JC: Cloning of the
cDNA's encoding the cellular retinaldehyde binding protein
from bovine and human retina and comparison of the protein
sequences. J Bioi Chem 1988,263: 18688-702.
24. Yamaki K, Tsuda M, Shinohara T: The sequence of human
retinal S-antigen reveals similarities with alpha-transducin.
FEBSLett 1988,234: 39-43.
25. Lerea CL, Bunt-Milam AH, Hurley JB: Alpha transducin is
present in blue-, green-, and red-sensitive cone photorecep­
tors in the human retina. Neuron1989,3: 367-76.
26. Ovchinnikov YA, Lipkin V M, Kumarev VP, Gubanov V V,
Khramtsov NV, Akhmedov NB, Zagranichny V E, Muradov
KG: Cyclic GMP phosphodiesterase from cattle retina.
Amino acid sequence of the gamma-subunit and nucleotide
sequence of the corresponding cDNA. FEBSLett1986,204:
288-92.
27. Tuteja N and Farber DB: Gamma-subunit of mouse retinal
cyclic-GMP phosphodiesterase cDNA and corresponding
amino acid sequence. FEBS Lett 1988,232: 182-6.
28. Ringens PJ, Fang M, Shinohara T, Bridges CD, Lerea CL,
Berson EL, Dryja TP: Analysis of genes coding for S-anti­
gen, interstitial retinol binding protein, and the alpha-sub­
unit of cone transducin in patients with retinitis pigmentosa.
Invest Ophthal Vis Sci 1990,31: 1421-6.
29. Cotran PR, Ringens PJ, Crabb JW, Berson EL, Dryja TP.
Analysis of the DNA of patients with retinitis pigmentosa
with a cellular retinaldehyde binding protein cDNA. Exp
EyeRes 1990,51: 15-19.
30. Cotran PR, Bruns GAP, Berson EL, Dryja TP: Genetic anal­
ysis of patients with retinitis pigmentosa using a cloned
cDNA probe for the human gamma subunit of cyclic GMP
phosphodiesterase. Exp Eye Res 1991,53: 557-64.
31. Liou GI, Li Y, Wang C, Fong SL, Bhattacharya S, Bridges
CDB: BglII RFLP recognised by a human IRBP cDNA
localised to chromosome 10. Nuc AcidsRes1987,15: 3196.
32. Simpson NE, Kidd KK, Goodfellow PJ, Dermid HM, Myer
S, Kidd JR, Jackson CE, Duncan AMV, Farrer LA, Brasch
K, Castiglione C, Genel M, Gertner J, Greenberg CR, Gus­
ella JF, Holden JJA, White BN: Assignment of multiple
endocrine neoplasia type 2A to chromosome 10 by linkage.
Nature1987,328: 528-30.
33. Nathans J, Piantanida TP, Eddy RL, Shows T B, Hogness
DS: Molecular genetics of inherited variation in human
colour vision. Science 1986,232: 203-10.
34. Sparkes RS, Klisak I, Kaufman D, Mohandas T, Tobin AJ,
McGinnis J: Assignment of the rhodopsin gene to human
chromosome three, region 3q21-3q24 by in situ hybrid­
isation studies. CurrEye Res 1986,5: 797-8.
35. Yandell DW, Campbell TA, Dayton SH, Petersen R, Walton
D, Little JB, McConkie-Rosell A, Buckley EG, Dryja TP:
Identification of oncogenic point mutations in the human
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
9
retinoblastoma gene and application to genetic counseling.
NEng J Med 1989,321: 1689-95.
Yandell DW and Dryja TP: Direct genomic sequence of
alleles at the human retinoblastoma locus: application to
cancer diagnosis and gentic counseling. In: Furth M,
Greaves M, eds. Cold Spring Harbour Symposium Series:
Cancer Cells 7-Molecular Diagnostics of Human Cancer.
Cold Spring Harbor: Cold Spring Harbour Press, 1989:
223-227.
Dryja TP, McGee T L, Reichel E, Hahn LB, Cowley GS,
Yandell DW, Sandberg MA, Berson EL: A point mutation of
the rhodopsin gene in one form of retinitis pigmentosa.
Nature 1990,343: 364-6.
Applebury ML and Hargrave PA: Molecular biology of the
visual pigments. Vision Res 1986, 12: 1881-95.
Baehr W, Falk JD, Bugra K, Triantafyllos JT, McGinnis JF:
Isolation and analysis of the mouse opsin gene. FEBS Lett
1988,238: 253-6.
Pappin DJC, Eliopoulos E, Brett M, Findlay JBC: A struc­
tural model for ovine rhodopsin. IntJ Bioi Macromol1984,
6: 73-6.
Weber JL and May PE: Abundant class of human DNA
polymorphisms which can be typed using the polymerase
chain reaction. AmJ Hum Genet 1989,44: 388-96.
Dryja TP, McGee TL, Hahn LB, Cowley GS, Olsson JE,
Reichel E, Sandberg MA, Berson EL: Mutations within the
rhodopsin gene in patients with autosomal dominant retinitis
pigmentosa. NEng J Med 1990,323: 1302-7.
Bhattacharya S, Lester D, Keen T J, Bashir R, Lauffart B,
Inglehearn CF, Jay M, Bird AC: Retinitis pigmentosa and
mutations in rhodopsin. Lancet 1991,337: 185.
Fujiki K, Hotta Y, Shiono T, Hayakawa M, Noro M, Sakuma
T, Tarnai M, Nakajima A, Kanai A: Codon 347 mutation of
the rhodopsin gene in a Japanese family with autosomal
dominant retinitis pigmentosa. Am J Hum Genet 1991,
49(Suppl): 187.
Doi T, Molday RS, Khorana HG: Role of the intradiscal
domain in rhodopsin assembly and function. ProcNat! Acad
Sci USA 1990,87: 4991-5.
Papermaster DS, Dreyer WJ: Rhodopsin content in the outer
segment membranes of bovine and frog retinal rods. Bio­
chem 1974,13: 2438-44.
Young RW: Visual cells and the concept of renewal. Invest
Ophthalmol Vis Sci 1976,15: 700-25.
Fliesler SJ, Rayborn ME, Hollyfield JG: Membrane mor­
phogenesis in retinal rod outer segments: inhibition by tu­
nicamycin. J Cell Bioi 1985, 100: 574-87.
Ulshafer RJ, Allen CB, Fliesler SJ: Tunicamycin-induced
dysgenesis of retinal rod outer segment membranes. Invest
Ophthal Vis Sci 1986,27: 1587-601.
Sheffield V C, Fishman GA, Beck JS, Kimura AE, Stone
EM: Identification of novel rhodopsin mutations associated
with retinitis pigmentosa by GC-clamped denaturing gradi­
ent gel electrophoresis. Am J Hum Genet 1991, 49:
699-706.
Dryja TP, Hahn LB, Cowley GS, McGee T L, Berson EL:
Mutation spectrum of the rhodopsin gene among patients
with autosomal dominant retinitis pigmentosa. Proc Nat!
Acad Sci USA 1991,88: 9370-4.
Sung CH, Davenport CM, Hennessey JC, Maumenee IH,
Jacobson SG, Heckenlively JR, Nowakowski R, Fishman G,
Gouras P, Nathans J: Rhodopsin mutations in autosomal
dominant retinitis pigmentosa. Proc Natl Acad Sci USA
1991,88: 6481-5.
Sung CH, Schneider BG, Agarwal N, Papermaster DS,
Nathans J: Functional heterogeneity of mutant rhodopsin
responsible for autosomal dominant retinitis pigmentosa.
Proc Natl Acad Sci USA 1991,88: 8840-4.
Berson EL, Rosner B, Sandberg MA, Dryja TP: Ocular find­
ings in patients with autosomal dominant retinitis pigmen-
10
T. P. DRYJA
tosa and a rhodopsin gene defect (Pro23His). Arch
Ophthalmol1991,109: 92-101.
55. Curcio CA, Sloan KR, Kalina RE, Hendrickson AE: Human
6 2. Stone EM, Kimura AE, Nichols BE, Khadivi P, Fishman
photoreceptor topography. J Comp Neurol 1990, 292:
497-523.
Berson EL, Rosner B, Sandberg MA, Weigel-DiFranco C,
Dryja TP: Ocular findings in patients with autosomal domi­
nant retinitis pigmentosa and rhodopsin, proline-347-leu­
cine. AmJ Ophthalmol1991,111: 614-23.
Jacobson SG, Kemp CM, Sung CH, Nathans J: Retinal func­
tion and rhodopsin levels in autosomal dominant retinitis
pigmentosa with rhodopsin mutations. Am J Ophthalmol
1991,112: 256-71.
Fishman GA, Stone EM, Gilbert LD, Kenna P, Sheffield
V C: Ocular findings associated with a rhodopsin gene codon
58 tranversion mutation in autosomal dominant retinitis pig­
mentosa. Arch Ophthalmol199 1,109: 1387-93.
Berson EL, Sandberg MA, Dryja TP: Autosomal dominant
retinitis pigmentosa with rhodopsin, valine-345-methio­
nine. Trans Am Ophthalmol Soc1991; (in press).
Heckenlively JR, Rodriguez JA, Daiger SP: Autosomal
dominant sectoral retinitis pigmentosa. Two families with
transversion mutation in codon 23 of rhodopsin. Arch Oph­
thalmol1991,109: 84-91.
Richards JE, Kuo C Y, Boehnke M, Sieving PA: Rhodopsin
Thr58Arg mutation in a family with autosomal dominant
retinitis pigmentosa. Ophthalmolog y 1991,98: 1797-805.
codon 23 of the rhodopsin gene. Ophthalmolog y 1991,98:
56.
57.
58.
59.
60.
61.
GA, Sheffield VC: Regional distribution of retinal degener­
ation in patients with the proline to histidine mutation in
1806-13.
63. Sorscher EJ and Huang Z: Diagnosis of genetic disease by
primer-specified restriction map modification, with appli­
cation to cystic fibrosis and retinitis pigmentosa. Lancet
1991,337: 1115-18.
64. Farrar GJ, Kenna P, Redmond R, Shiels D, McWilliam P,
Humphries MM, Sharp EM, Jordan S, Kumar-Singh R,
Humphries P: Autosomal dominant retinitis pigmentosa: a
mutation in codon 178 of the rhodopsin gene in two families
of Celtic origin. Genomics 199 1,11: 1170-1.
65. Keen TJ, Ingleheam CF, Lester DH, Bashir R, Jay M, Bird
AC, Jay B, Bhattachary SS: Autosomal dominant retinitis
pigmentosa: four new mutations in rhodopsin, one of them
in the retinal attachment site. Genomics1991,11: 199-205.
66. Inghleheam CF, Bashir R, Lester DH, Jay M, Bird AC, Bhat­
tacharya SS: A 3-bp deletion in the rhodopsin gene in a
family with autosomal dominant retinitis pigmentosa. AmJ
Hum Genet 1991,48: 26-30.
67. Gal A, Artlich A, Ludwig M, Niemeyer G, Olek K, Schwin­
ger E, Schinzel A: Pro347Arg mutation of the rhodopsin
gene in autosomal dominant retinitis pigmentosa. Genomics
1991,11: 468-70.